how to justify a sample size for an underpowered pilot study

Status
Not open for further replies.
#1
moved to biostatics: how to justify a sample size for an underpowered pilot study

UPDATE: I reposted this in the biostatistics forum. I also added some more details, mainly my not-fully-formed thoughts about how to proceed. Thanks!

I need to justify a sample size of 30 or 32 (15 or 16 per arm) for a placebo-controlled pilot clinical trial of a drug. This is all I can afford, but that is not a good enough justification for grant reviewers. Here is all the info I have. It is all from previously published studies:

In a prior study of 50 people, 24 got this drug and 26 got placebo. On average, the drug group declined 0.54 points while the placebo group declined 6.9 points (less of a decline is good.)
The difference between drug & placebo was significant at p=.015 (presumably 2-tailed).
I don't have standard deviation.
A 2 point change is considered a minimally clinically significant difference on this outcome measure.

Using online calculators and inputting p values, I estimated the effect size at about .634.

Referencing this article if needed, how can I justify a sample size of 30 or 32 even though it is underpowered for hypothesis testing?

From Lee, 2014 The statistical interpretation of pilot trials: should significance thresholds be reconsidered?:
“We recommend that in pilot trials the focus should be on descriptive statistics and estimation, using confidence intervals, rather than formal hypothesis testing. We further recommend that confidence intervals in addition to 95% confidence intervals, such as 85% or 75%, be used for the estimation. The confidence interval should then be interpreted with regards to the minimum clinically important difference and we suggest setting minimum prior requirements.”
 
Last edited:
Status
Not open for further replies.